Difference in SimSolid vs Optistruct result

Junta_21946

Hi expert,

I'm using Inspire2022 to perform a structural analysis of a simple steel assembly.

While using Inspire, I have tried both Simsolid and OptiStruct as the solver.

With a small assembly or a single part, the result from 2 solvers are very close.

However, when working with some large assemblies, Optistruct gives different results compared to Simsolid.

Take an example below:

This is a bridge model from the Simsolid tutorial, I added it into Inspire (attached model) and run it with Simsolid and OS.

Boundary condition: fix 2 ends, 100kg mass on the middle bars with gravity on.

The displacement and structural behavior is very different (below image)

 (Optistruct)

 (Simsolid)

Can anyone explain the reason for this? And which result should be correct?

Note: you can use attach inspire model, switch solver to Optistruct >> setup the contact again then run.
(the Optistruct model is too large and cannot be attached)

I tried with Ansys. It gives a result similar to Optistruct  

Thank you!

Adriano_Koga

Try running SimSolid modelo with the  'Adapt for Stress'.

This will give you a more accurate model.

"for Stiffness" is quite quick for running a model, but should be used more for initial evaluation, and validating contacts and connections.

For final run you would probably want to use more passes and feature recognition.

 

In SimSolid you should probably perform a kind of 'convergence study' using the 'custom' option, and changing number of passes (default is 3) and adapt to features and adapt to thin solids.

For some structures this can enhance a lot your response.

So, run with different options and compare maximum displacements and stresses. At a certain point they won't change too much.

 

Below the result using SimSolid solver with adapt for stress, matching quite well OptiStruct results, in less than 15seconds.